CN104636750B - A kind of pavement crack recognizer and system based on double scale clustering algorithms - Google Patents
A kind of pavement crack recognizer and system based on double scale clustering algorithms Download PDFInfo
- Publication number
- CN104636750B CN104636750B CN201410787530.5A CN201410787530A CN104636750B CN 104636750 B CN104636750 B CN 104636750B CN 201410787530 A CN201410787530 A CN 201410787530A CN 104636750 B CN104636750 B CN 104636750B
- Authority
- CN
- China
- Prior art keywords
- crack
- minimum
- crack area
- area
- clustered
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/285—Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of pavement crack recognizers and system based on double scales cluster：Computer reads 3 d image data matrix, obtains binary image；According to sequence from top to bottom, from left to right, using the corresponding data matrix of eight neighborhood searching algorithm scanning binary image, the crack area after being marked；In the corresponding ellipse of crack area, the crack area after crack is clustered is carried out using double scale clustering algorithms；It is external oval as pavement crack using the minimum where the crack area after cluster.Complexity of the present invention is low, run time is short, is participated in without artificial.By mixed and disorderly crack data local using the thought characterization of linear fit, model construction as mathematic(al) representation that is regular, determining, to the complexity of the data processing reduced；The detection that pavement crack can be completed in collected pavement crack data need to be only inputted, therefore the algorithm detection efficiency is high, speed is fast, has certain researching value.
Description
Technical field
This patent belongs to field of road, particularly relates to a kind of pavement crack identification calculation based on double scale clustering algorithms
Method.
Background technology
Traditional pavement crack identification technology is all that crack of linking closely carries out a series of image procossing and constantly makes its effect
It continues to optimize, is convenient for Objective extraction, i.e., such algorithm is directed generally in the extraction of FRACTURE CHARACTERISTICS, and seldom considers to be actually needed
And whether treated pavement crack really corresponds to the practical pavement crack of same.First, traditional FRACTURE CHARACTERISTICS extraction
The practical road surface of diminution that algorithm makes pretreated pavement crack binary map always different degrees of for original fracture
Crack area is broken what is more after repeatedly handling so that phenomenon of rupture is presented in same pavement crack on the image in practice
If the crack after splitting very likely is taken as two even more crack treatments without timely finding and repairing,
Since pavement crack identification process is a serial processing procedure, there are Accumulation Phenomenons for mistake, so that subsequent processing
The results of the work based on front end mistake such as middle crack positioning carry out, and necessarily lead to the crack identification of mistake as a result, substantially increasing
That is, there is the phenomenon that " not mending then wrong " in the error rate of crack identification.Second, for reality, the purpose of pavement crack detection exists
In Accurate classification crack, the positioning of crack and crack actual area is accurately positioned, to be provided reliably for highway maintenance department
Data, in favor of its carry out highway maintenance management.
Clustering algorithm has great application prospect for image segmentation field.It not only has in terms of handling mass data
There is prodigious advantage, and there is excellent scalability, new research method is found convenient for upper from different angles.Different is poly-
Class algorithm presses the difference of its clustering criteria, can be divided into " hard " cluster, and " soft " cluster.For simple " hard " cluster, collection is enabled
It closes C and indicates gray value of image data set, clustering is carried out to it be equivalent to that it is divided into subregion c by certain criterion1,
c2.......ck, k is classification number.So that subregion meets condition：Non-empty：；Integrality：c1∪c2∪c3∪…
∪ck=C, clustering algorithm are substantially the reallocation to initial data, by mining data internal structure, are constantly looked for more excellent
The clustering algorithm of change, so that the data after reallocating embody certain internal consistency, the embodiment of this consistency is usually again
It is weighed by specific criterion function, will obtain different using different criterion functions as a result, Optimality Criteria function is optimization
One direction of clustering algorithm.Common clustering algorithm has hierarchical clustering algorithm, mixing interpretive model search algorithm, nearest-neighbor
Clustering algorithm, fuzzy clustering algorithm, artificial neural network clustering algorithm, genetic algorithm for clustering etc..These clustering algorithms are made a general survey of,
Core is the expression of " distance ", and " distance " under different criterion embodies different Clustering Effects, certainly, for different data
Set chooses different criterion and can be only achieved ideal effect accordingly, and therefore, the definition of " distance " is particularly critical.Together
When, with regard to pavement crack detection for, clustering algorithm the field the application that rarely has, and it is only should have also only be limited to road surface
The segmentation of crack image, there is no the detections in terms of progress crack area positioning.
Invention content
For the above pavement crack identification technology there are the problem of, the present invention proposes the pavement crack that is clustered based on double scales
Recognizer, using the thought of " first divide and gather afterwards ", i.e., first crack carries out pocket and divides part research, then using in optimization
Double scale clustering criterias of heart distance and differential seat angle carry out double scale clusters to fritter fault and fissures, finally use minimum external
Model of ellipse characterizes crack, realizes the positioning in crack and defining for region, reaches the identification of pavement crack.Not only avoiding makes
The data volume caused by calculating manifold distance is big, the high realization brought of complexity is difficult, and has reached and clustered using manifold distance
The advantages of bringing.
In order to achieve the above object, the present invention adopts the following technical scheme that：
A kind of pavement crack recognizer based on double scales cluster, includes the following steps：
Step 1：Computer reads 3 d image data matrix, is filtered and is filtered to 3 d image data matrix
3 d image data matrix after wave is then converted into gray level image, and is carried out using Otsu algorithms to the gray level image
Binary conversion treatment obtains binary image；
Step 2：According to sequence from top to bottom, from left to right, two obtained using eight neighborhood searching algorithm scanning step 1
The corresponding data matrix of value image, the crack area after being marked, and minimum external ellipse is used to each crack area
It is characterized (use minimum external ellipse by crack area circle in it)；
Step 3：In the corresponding ellipse of crack area, crack cluster is carried out using double scale clustering algorithms, after obtaining cluster
Crack area；
Step 4：It is external oval as pavement crack using the minimum where the crack area after cluster.
Further, which is characterized in that the step 2 specifically comprises the following steps：
Step 21：According to the filtered 3 d image data matrix of sequential scan from top to bottom, from left to right；Successively with
Centered on each data point, judge whether the eight neighborhood of central point is crack point, if it is, the point within the scope of eight neighborhood
Belong to the basic unit region put centered on the data point, and each basic unit region sequentially numbered and (is marked),
Each basic unit region is used as crack area；
Step 22：Linear fit is carried out to all data points in the crack area of number i, obtains fitting a straight lineWherein, i=1,2,3 ...；niThe number that crack area for number i includes
Strong point number；The line segment length that the crack area intercepts in corresponding fitting a straight line is found out, a is denoted asi；Then the crack is calculated
The maximum value of distance is denoted as b by all data points to the distance of corresponding fitting a straight line in regioni, calculate aiWith horizontal direction
Angle is denoted as θi；
Step 23：By the parameter a of each crack areai、biAnd θiRespectively as corresponding elliptical long axis, short axle and
Deflection angle obtains the corresponding external ellipse of minimum of each crack area.
Further, the step 3 specifically comprises the following steps：
Step 31：For all slits region that step 2 obtains, choose long in the external ellipse of the corresponding minimum of crack area
The longest elliptical center of axis is as current cluster centre；
Step 32：It chooses and numbers minimum split in all slits region except the corresponding crack area of current cluster centre
The external ellipse of minimum where region is stitched, as target to be clustered；
Step 33：The center of target to be clustered is calculated to the distance of current cluster centre and the horizontal sextant angle at two centers
Difference；
Step 34：Whether the distance and horizontal sextant angle difference that judgment step 33 obtains meet following criterion function, if discontented
Foot, then go to step 37, if it is satisfied, then executing step 35；
Criterion function：J=(a0＜ O0Oi)&&(Δθoi＜ δ)
Wherein, a0For the representative external elliptical long axial length of minimum of current cluster centre, O0OiFor target to be clustered
Center is to the distance of current cluster centre, Δ θoiPoor, the δ for target to be clustered and the horizontal sextant angle of current cluster centre1It is normal
Number, takes 0~45 °；
Step 35：The class target to be clustered being classified as where current cluster centre, all ellipses that will include in such
(i.e. will be including all oval external oval circles using a minimum) be characterized using a new external ellipse of minimum, are calculated
The new external elliptical center of minimum, long axis, short axle and the deflection angle, and using the new external elliptical center of minimum as
Current cluster centre；
Step 36：Judge whether also unclassified crack area, if so, then continuing to choose next lowest number representative
Minimum external oval be used as target to be clustered, and return to step 33；If no, executing step 38；
Step 37：Current cluster centre is classified as a new class, returns to 31, otherwise, executes step 38；
Step 38：Terminate.
Further, in the step 1,3 d image data matrix is filtered using median filtering algorithm,
Binary conversion treatment is carried out to gray level image and uses Otsu algorithms.
It is a further object of the invention to provide a kind of pavement crack identifying systems based on double scales cluster, this is
System includes being sequentially connected the module connect as follows：
Image binaryzation module, to complete following function：
Computer reads 3 d image data matrix, is filtered to obtain to 3 d image data matrix filtered
3 d image data matrix is then converted into gray level image, and carries out binaryzation using Otsu algorithms to the gray level image
Processing, obtains binary image；
Crack area mark module, to complete following function：
According to sequence from top to bottom, from left to right, the binary picture obtained using eight neighborhood searching algorithm scanning step 1
As corresponding data matrix, the crack area after being marked, and minimum external oval progress table is used to each crack area
Sign；
Crack cluster module, in the corresponding ellipse of crack area, crack cluster is carried out using double scale clustering algorithms,
Crack area after being clustered；
Pavement crack extraction module, to split the external ellipse of minimum where the crack area after cluster as road surface
Seam.
Further, the crack area mark module is realizing the function of following below scheme：
Step 21：According to the filtered 3 d image data matrix of sequential scan from top to bottom, from left to right；Successively with
Centered on each data point, judge whether the eight neighborhood of central point is crack point, if it is, the point within the scope of eight neighborhood
Belong to the basic unit region put centered on the data point, and each basic unit region sequentially numbered and (is marked),
Each basic unit region is used as crack area；
Step 22：Linear fit is carried out to all data points in the crack area of number i, obtains fitting a straight lineWherein, i=1,2,3 ...；niThe number that crack area for number i includes
Strong point number；The line segment length that the crack area intercepts in corresponding fitting a straight line is found out, a is denoted asi；Then the crack is calculated
The maximum value of distance is denoted as b by all data points to the distance of corresponding fitting a straight line in regioni, calculate aiWith horizontal direction
Angle is denoted as θi；
Step 23：By the parameter a of each crack areai、biAnd θiRespectively as corresponding elliptical long axis, short axle and
Deflection angle obtains the corresponding external ellipse of minimum of each crack area.
Further, the crack cluster module is realizing the function of following flow：
Step 31：For all slits region that step 2 obtains, choose long in the external ellipse of the corresponding minimum of crack area
The longest elliptical center of axis is as current cluster centre；
Step 32：It chooses and numbers minimum split in all slits region except the corresponding crack area of current cluster centre
The external ellipse of minimum where region is stitched, as target to be clustered；
Step 33：The center of target to be clustered is calculated to the distance of current cluster centre and the horizontal sextant angle at two centers
Difference；
Step 34：Whether the distance and horizontal sextant angle difference that judgment step 33 obtains meet following criterion function, if discontented
Foot, then go to step 37, if it is satisfied, then executing step 35；
Criterion function：J=(a0＜ O0Oi)&&(Δθoi＜ δ)
Wherein, a0For the representative external elliptical long axial length of minimum of current cluster centre, O0OiFor target to be clustered
Center is to the distance of current cluster centre, Δ θoiPoor, the δ for target to be clustered and the horizontal sextant angle of current cluster centre1It is normal
Number, takes 0~45 °；
Step 35：The class target to be clustered being classified as where current cluster centre, all ellipses that will include in such
(i.e. will be including all oval external oval circles using a minimum) be characterized using a new external ellipse of minimum, are calculated
The new external elliptical center of minimum, long axis, short axle and the deflection angle, and using the new external elliptical center of minimum as
Current cluster centre；
Step 36：Judge whether also unclassified crack area, if so, then continuing to choose next lowest number representative
Minimum external oval be used as target to be clustered, and return to step 33；If no, executing step 38；
Step 37：Current cluster centre is classified as a new class, returns to 31, otherwise, executes step 38；
Step 38：Terminate.
Further, in described image binarization block, 3 d image data matrix is filtered using intermediate value
Filtering algorithm carries out binary conversion treatment to gray level image and uses Otsu algorithms.
Pavement crack identification technology proposed by the present invention has the following advantages that
1, the detection that pavement crack can be completed in collected pavement crack data need to be only inputted, therefore algorithm calculates letter
List, run time are short, are suitble to use in real-time system.
2, be not necessarily to it is artificial participate in, overcome that the labor intensity that man made ground's Crack Detection has is big, transplantability is poor, work effect
Rate is low and the poor disadvantage of filter effect.
3, on the basis of based on digital image processing techniques, the unique advantage of big data is handled using clustering algorithm, from
Completely new visual angle comprehensive analysis pavement crack image completes double scale clustering algorithms identification in crack.
Rambling pavement crack region is expressed as mathematics public affairs by the 4, method by establishing crack area mathematical model
The formula that formula can be indicated specifically is easy to use mathematical technique analysis.Strong Informational support is provided for the maintenance management on road surface,
Improve highway maintenance and management level.
Description of the drawings
Fig. 1 is the flow chart for the pavement crack recognizer of the present invention clustered based on double scales.
Fig. 2 is the result being respectively processed using three kinds of cracks of the present invention couple.Wherein, (a), (d), (g) are respectively original
Beginning transverse crack, original longitudinal crack and original chicken-wire cracking；(b), it is basic to be followed successively by (a), (d), the crack of (g) by (e), (h)
Unit；(c), (f), (i) are followed successively by the result after the algorithm process of (a), (d), (g) through the invention.
Fig. 3 is the schematic diagram for the pavement crack identifying system of the present invention clustered based on double scales.
Illustrate the implementation result of the pavement crack identification technology of the present invention below in conjunction with example：
Specific implementation mode
Referring to Fig. 1-Fig. 3, the pavement crack recognizer of the invention based on double scale clustering algorithms specifically includes as follows
Step：
Step 1：Computer read 3 d image data matrix, to 3 d image data matrix using median filtering algorithm into
Row is filtered, and is obtained filtered 3 d image data matrix, is then converted into gray level image, and to the gray level image
Binary conversion treatment is carried out using Otsu algorithms, obtains binary image；
It is low to the smoothness requirement of data using median filtering algorithm in the step, therefore it is suitble to faulting of slab ends in the present invention
The processing of data, to improve arithmetic speed.Meanwhile using Otsu algorithms being a kind of adaptive thresholding algorithm, automate journey
Degree is high.
Step 2：According to sequence from top to bottom, from left to right, two obtained using eight neighborhood searching algorithm scanning step 1
The corresponding data matrix of value image, the crack area after being marked, and minimum external ellipse is used to each crack area
It is characterized (use minimum external ellipse by crack area circle in it)；
Specifically comprise the following steps：
Step 21：According to the filtered 3 d image data matrix of sequential scan from top to bottom, from left to right；Successively with
Centered on each data point, judge whether the eight neighborhood of central point is crack point, if it is, the point within the scope of eight neighborhood
Belong to the basic unit region put centered on the data point, and each basic unit region sequentially numbered and (is marked),
Each basic unit region is used as crack area；
Step 22：Linear fit is carried out to all data points in the crack area of number i, obtains fitting a straight lineWherein, i=1,2,3 ...；niThe number that crack area for number i includes
Strong point number；The line segment length that the crack area intercepts in corresponding fitting a straight line is found out, a is denoted asi；Then the crack is calculated
The maximum value of distance is denoted as b by all data points to the distance of corresponding fitting a straight line in regioni, calculate aiWith horizontal direction
Angle is denoted as θi；
Step 23：By the parameter a of each crack areai、biAnd θiRespectively as corresponding elliptical long axis, short axle and
Deflection angle obtains the corresponding external ellipse of minimum of each crack area.
In the step, using eight neighborhood searching algorithm, search is comprehensive so that result is more accurate；Meanwhile using minimum external
Ellipse is characterized more existing rectangle characterization, and the characterization of fracture is more accurate.
Step 3：In the corresponding ellipse of crack area, crack cluster is carried out using double scale clustering algorithms, after obtaining cluster
Crack area；Specifically comprise the following steps：
Step 31：For all slits region that step 2 obtains, choose long in the external ellipse of the corresponding minimum of crack area
The longest elliptical center of axis is as current cluster centre；
Step 32：It chooses and numbers minimum split in all slits region except the corresponding crack area of current cluster centre
The external ellipse of minimum where region is stitched, as target to be clustered；
Step 33：The center of target to be clustered is calculated to the distance of current cluster centre and the horizontal sextant angle at two centers
Difference；
Step 34：Whether the distance and horizontal sextant angle difference that judgment step 33 obtains meet following criterion function, if discontented
Foot, then go to step 37, if it is satisfied, then executing step 35；
Criterion function：J=(a0＜ O0Oi)&&(Δθoi＜ δ)
Wherein, a0For the representative external elliptical long axial length of minimum of current cluster centre, O0OiFor target to be clustered
Center is to the distance of current cluster centre, Δ θoiPoor, the δ for target to be clustered and the horizontal sextant angle of current cluster centre1It is normal
Number, takes 0~45 °；
Step 35：The class target to be clustered being classified as where current cluster centre, all ellipses that will include in such
(i.e. will be including all oval external oval circles using a minimum) be characterized using a new external ellipse of minimum, are calculated
The new external elliptical center of minimum, long axis, short axle and the deflection angle, and using the new external elliptical center of minimum as
Current cluster centre；
Step 36：Judge whether also unclassified crack area, if so, then continuing to choose next lowest number representative
Minimum external oval be used as target to be clustered, and return to step 33；If no, executing step 38；
Step 37：Current cluster centre is classified as a new class, returns to 31, otherwise, executes step 38；
Step 38：Terminate.
The selection of criterion function in the step, can judge distance again can judge angle, can comprehensive error in judgement, improve
As a result precision.
Step 4：The external ellipse of minimum where using the crack area after the cluster obtained in step 3 is split as road surface
Seam, realizes the detection and localization of pavement crack.
Referring to Fig. 2, to use double scale clustering algorithms to identify pavement crack as a result, δ=30 ° are chosen in experiment.As it can be seen that using
Double scale clustering algorithms can obtain accurate crack range, crack information extraction, to complete the detection in crack.
Claims (4)
1. a kind of pavement crack recognition methods based on double scales cluster, which is characterized in that include the following steps：
Step 1：Computer reads 3 d image data matrix, is filtered after obtaining filtering to 3 d image data matrix
3 d image data matrix, be then converted into gray level image, and two-value is carried out using Otsu algorithms to the gray level image
Change is handled, and obtains binary image；
Step 2：According to sequence from top to bottom, from left to right, the binaryzation obtained using eight neighborhood searching algorithm scanning step 1
The corresponding data matrix of image, the crack area after being marked, and minimum external oval progress is used to each crack area
Characterization；The step 2 specifically comprises the following steps：
Step 21：According to the corresponding data matrix of the filtered binary image of sequential scan from top to bottom, from left to right；According to
It is secondary centered on each data point, judge whether the eight neighborhood of central point is crack point, if it is, within the scope of eight neighborhood
Point belong to the basic unit region put centered on the data point, and each basic unit region is sequentially numbered, Mei Geji
This unit area is used as crack area；
Step 22：Linear fit is carried out to all data points in the crack area of number i, obtains fitting a straight lineWherein, i=1,2,3 ...；niThe number that crack area for number i includes
Strong point number；The line segment length that the crack area intercepts in corresponding fitting a straight line is found out, a is denoted asi；Then the crack is calculated
The maximum value of distance is denoted as b by all data points to the distance of corresponding fitting a straight line in regioni, calculate aiWith horizontal direction
Angle is denoted as θi；
Step 23：By the parameter a of each crack areai、biAnd θiRespectively as corresponding elliptical long axis, short axle and deflection
Angle obtains the corresponding external ellipse of minimum of each crack area；
Step 3：In the corresponding ellipse of crack area, crack cluster, splitting after being clustered are carried out using double scale clustering algorithms
Stitch region；The step 3 specifically comprises the following steps：
Step 31：For all slits region that step 2 obtains, long axis is chosen in the external ellipse of corresponding minimum of crack area most
Long elliptical center is as current cluster centre；
Step 32：It chooses and numbers minimum crack area in all slits region except the corresponding crack area of current cluster centre
The external ellipse of minimum where domain, as target to be clustered；
Step 33：The center for calculating target to be clustered is poor to the distance of current cluster centre and the horizontal sextant angle at two centers；
Step 34：Whether the distance and horizontal sextant angle difference that judgment step 33 obtains meet following criterion function, if not satisfied, then
Step 37 is gone to, if it is satisfied, then executing step 35；
Criterion function：J=(a0<O0Oi)&&(Δθoi<δ1)
Wherein, a0For the representative external elliptical long axial length of minimum of current cluster centre, O0OiFor the center of target to be clustered
To the distance of current cluster centre, Δ θoiPoor, the δ for target to be clustered and the horizontal sextant angle of current cluster centre1For normal number, take
0~45 °；
Step 35：The class target to be clustered being classified as where current cluster centre, all oval uses that will include in such
One new external ellipse of minimum is characterized, and the new external elliptical center of minimum, long axis, short axle and the deflection is calculated
Angle, and using the new external elliptical center of minimum as current cluster centre；
Step 36：Judge whether also unclassified crack area, if so, then continuing to choose next lowest number representative most
Small external ellipse is used as target to be clustered, and return to step 33；If no, executing step 38；
Step 37：Current cluster centre is classified as a new class, returns to 31, otherwise, executes step 38；
Step 38：Terminate；
Step 4：It is external oval as pavement crack using the minimum where the crack area after cluster.
2. the pavement crack recognition methods as described in claim 1 based on double scales cluster, which is characterized in that the step 1
In, 3 d image data matrix is filtered using median filtering algorithm.
3. a kind of pavement crack identifying system based on double scales cluster, which is characterized in that including being sequentially connected the mould connect as follows
Block：
Image binaryzation module, to complete following function：
Computer reads 3 d image data matrix, is filtered to obtain filtered three-dimensional to 3 d image data matrix
Image data matrix is then converted into gray level image, and carries out binary conversion treatment using Otsu algorithms to the gray level image,
Obtain binary image；
Crack area mark module, to complete following function：
According to sequence from top to bottom, from left to right, two obtained using eight neighborhood searching algorithm scan image binarization block
The corresponding data matrix of value image, the crack area after being marked, and minimum external ellipse is used to each crack area
It is characterized；
The function of the crack area mark module specific implementation following below scheme：
Step 21：According to the corresponding data matrix of the filtered binary image of sequential scan from top to bottom, from left to right；According to
It is secondary centered on each data point, judge whether the eight neighborhood of central point is crack point, if it is, within the scope of eight neighborhood
Point belong to the basic unit region put centered on the data point, and each basic unit region is sequentially numbered, Mei Geji
This unit area is used as crack area；
Step 22：Linear fit is carried out to all data points in the crack area of number i, obtains fitting a straight lineWherein, i=1,2,3 ...；niThe number that crack area for number i includes
Strong point number；The line segment length that the crack area intercepts in corresponding fitting a straight line is found out, a is denoted asi；Then the crack is calculated
The maximum value of distance is denoted as b by all data points to the distance of corresponding fitting a straight line in regioni, calculate aiWith horizontal direction
Angle is denoted as θi；
Step 23：By the parameter a of each crack areai、biAnd θiRespectively as corresponding elliptical long axis, short axle and deflection
Angle obtains the corresponding external ellipse of minimum of each crack area；
Crack cluster module, in the corresponding ellipse of crack area, to carry out crack cluster using double scale clustering algorithms, obtain
Crack area after cluster；The crack cluster module is realizing the function of following flow：
Step 31：For all slits region that step 2 obtains, long axis is chosen in the external ellipse of corresponding minimum of crack area most
Long elliptical center is as current cluster centre；
Step 32：It chooses and numbers minimum crack area in all slits region except the corresponding crack area of current cluster centre
The external ellipse of minimum where domain, as target to be clustered；
Step 33：The center for calculating target to be clustered is poor to the distance of current cluster centre and the horizontal sextant angle at two centers；
Step 34：Whether the distance and horizontal sextant angle difference that judgment step 33 obtains meet following criterion function, if not satisfied, then
Step 37 is gone to, if it is satisfied, then executing step 35；
Criterion function：J=(a0<O0Oi)&&(Δθoi<δ1)
Wherein, a0For the representative external elliptical long axial length of minimum of current cluster centre, O0OiFor the center of target to be clustered
To the distance of current cluster centre, Δ θoiPoor, the δ for target to be clustered and the horizontal sextant angle of current cluster centre1For normal number, take
0~45 °；
Step 35：The class target to be clustered being classified as where current cluster centre, all oval uses that will include in such
One new external ellipse of minimum is characterized, and the new external elliptical center of minimum, long axis, short axle and the deflection is calculated
Angle, and using the new external elliptical center of minimum as current cluster centre；
Step 36：Judge whether also unclassified crack area, if so, then continuing to choose next lowest number representative most
Small external ellipse is used as target to be clustered, and return to step 33；If no, executing step 38；
Step 37：Current cluster centre is classified as a new class, returns to 31, otherwise, executes step 38；
Step 38：Terminate；
Pavement crack extraction module oval is used as pavement crack to the minimum where the crack area after clustering is external.
4. the pavement crack identifying system as claimed in claim 3 based on double scales cluster, which is characterized in that described image two
In value module, 3 d image data matrix is filtered using median filtering algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410787530.5A CN104636750B (en) | 2014-12-17 | 2014-12-17 | A kind of pavement crack recognizer and system based on double scale clustering algorithms |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410787530.5A CN104636750B (en) | 2014-12-17 | 2014-12-17 | A kind of pavement crack recognizer and system based on double scale clustering algorithms |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104636750A CN104636750A (en) | 2015-05-20 |
| CN104636750B true CN104636750B (en) | 2018-09-21 |
Family
ID=53215478
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410787530.5A Expired - Fee Related CN104636750B (en) | 2014-12-17 | 2014-12-17 | A kind of pavement crack recognizer and system based on double scale clustering algorithms |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104636750B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105320942B (en) * | 2015-10-21 | 2018-07-06 | 东南大学 | A kind of road surface breakage detection method based on combined detector |
| CN105825169B (en) * | 2016-03-10 | 2019-02-15 | 辽宁工程技术大学 | A kind of pavement crack recognition methods based on road image |
| CN106204497B (en) * | 2016-07-20 | 2018-12-25 | 长安大学 | A kind of pavement crack extraction algorithm based on smooth smoothed curve and matched curve |
| CN107610094B (en) * | 2017-08-02 | 2020-04-03 | 长安大学 | Three-dimensional crack detection method based on ellipsoid three-dimensional representation |
| CN109839439B (en) * | 2019-03-21 | 2021-08-20 | 广西安盛建设工程检测咨询有限公司 | Road and bridge pavement crack detection system and detection method thereof |
| CN110334727B (en) * | 2019-04-30 | 2021-04-27 | 国家电网有限公司 | Intelligent matching detection method for tunnel cracks |
| CN110909628A (en) * | 2019-11-05 | 2020-03-24 | 长安大学 | Natural illumination compensation method for detecting pavement cracks with shadows |
| CN110853034A (en) * | 2020-01-15 | 2020-02-28 | 清华四川能源互联网研究院 | Crack detection method, crack detection device, electronic equipment and computer-readable storage medium |
| CN115424178B (en) * | 2022-09-05 | 2023-07-14 | 兰州大学 | Enhancement method for improving pavement crack data identification |
| CN118691609A (en) * | 2024-08-23 | 2024-09-24 | 中铁九局集团第一建设有限公司 | Expressway crack detection method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1563891A (en) * | 2004-04-20 | 2005-01-12 | 长安大学 | System and method for discriminating road gap |
| US8081798B2 (en) * | 2007-11-20 | 2011-12-20 | Lawrence Livermore National Security, Llc | Method and system for detecting polygon boundaries of structures in images as particle tracks through fields of corners and pixel gradients |
| CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
| CN103903268A (en) * | 2014-04-09 | 2014-07-02 | 南京理工大学 | Road surface crack detection method based on multiple lower-layer marked features |
-
2014
- 2014-12-17 CN CN201410787530.5A patent/CN104636750B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1563891A (en) * | 2004-04-20 | 2005-01-12 | 长安大学 | System and method for discriminating road gap |
| US8081798B2 (en) * | 2007-11-20 | 2011-12-20 | Lawrence Livermore National Security, Llc | Method and system for detecting polygon boundaries of structures in images as particle tracks through fields of corners and pixel gradients |
| CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
| CN103903268A (en) * | 2014-04-09 | 2014-07-02 | 南京理工大学 | Road surface crack detection method based on multiple lower-layer marked features |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636750A (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104636750B (en) | A kind of pavement crack recognizer and system based on double scale clustering algorithms | |
| WO2020233152A1 (en) | Urban building space data-based built-up area boundary identification method and device | |
| CN107527067B (en) | Railway roadbed disease intelligent identification method based on ground penetrating radar | |
| CN110349122A (en) | A kind of pavement crack recognition methods based on depth convolution fused neural network | |
| CN106803253B (en) | A kind of three-dimensional rock image crack identification method | |
| CN107680678A (en) | Based on multiple dimensioned convolutional neural networks Thyroid ultrasound image tubercle auto-check system | |
| CN109064461A (en) | A kind of detection method of surface flaw of steel rail based on deep learning network | |
| CN107016665A (en) | A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks | |
| CN105787486B (en) | A kind of girder steel crack detecting method based on image procossing | |
| CN106339707B (en) | A kind of gauge pointer image-recognizing method based on symmetric characteristics | |
| Wang et al. | Cracking classification using minimum rectangular cover–based support vector machine | |
| CN102645679A (en) | Mesocyclone identification method based on Doppler radar echo images | |
| CN109558902A (en) | A kind of fast target detection method | |
| CN112964712A (en) | Method for rapidly detecting state of asphalt pavement | |
| CN107392141A (en) | A kind of airport extracting method based on conspicuousness detection and LSD straight-line detections | |
| CN104268600B (en) | A kind of mineral flotation foam image texture analysis and operating mode's switch method based on Minkowski distances | |
| CN105825169A (en) | Road-image-based pavement crack identification method | |
| CN108830270A (en) | Correctly divide the localization method of the language of the Manchus word central axes of each identification to language of the Manchus word | |
| CN107492084B (en) | Typical clustering cell nucleus image synthesis method based on randomness | |
| CN114283407A (en) | Self-adaptive automatic leukocyte segmentation and subclass detection method and system | |
| CN110704559B (en) | Multi-scale vector surface data matching method | |
| CN104820826B (en) | A kind of domatic extraction and recognition methods based on digital elevation model | |
| CN104809756A (en) | Asphalt void space structure reconstruction method based on X-ray CT image | |
| CN104599321A (en) | Building method of discrete element module of true aggregate particles based on X-ray CT image | |
| CN109993773B (en) | Multi-target tracking method and device for series section images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180921 Termination date: 20201217 |